Introduction

This page showcase seven new features of enhanced visualization with ShinyCell2, outlined below. Please refer to the corresponding section for more detailed explanations of the features and the visual examples for demonstrations.

  1. Zoom-enabled UMAP
  2. Separate Legend Rendering
  3. Extreme Value Handling
  4. Side-by-side Visualization
  5. Statistical Testing on Violin/Box Plots
  6. Flexible Ordering on Proportion Plots
  7. Customized Color Scale on Bubble Plot/Heatmap

1. Zoom-enabled UMAP

For single-cell analysis, researchers are often interested in examining a specific region of the UMAP. The current enhanced feature enables users to zoom into a particular region of UMAP without loss of resolution. At the same time, a global and a local perspective of the UMAP is displayed simultaneously with dynamic displays of cell counts in both the full and zoomed-in views.

2. Separate Legend Rendering

To provide users with easy image post-processing, this current enhanced feature allows users to export main plots and legends as separate PNG or PDF files. More importantly, this feature ensures that the main plots (boxed red below) are always of the same consistent sizing regardless of the size of the legends (boxed blue below).

3. Extreme Value Handling

It is not uncommon to have extreme values i.e. outliers in the data that distort the color scale of the display, making it hard to interpret the inflated color range. This enhanced feature applies a percentile-based minimum and maximum cutoffs, effectively minimizing the influence of outliers and enhancing the clarity of color-based data representations. Here, we show an example below where there are some outlier values (circled blue on the left panel). After applying a q95 max cutoff (right panel), the updated colour scale better reflects the data distribution.

4. Side-by-side Visualization

In order to facilitate comparison across modalities, for example, simultaneously examining cell metadata and assay expressions, we set up both the “Side-by-side Spatial” and “Side-by-side DimRed” tabs where users can display the relevant information of interest in a side-by-side manner. Additional example includes display of confusion matrices to compare cell memberships between categorical variable pairs or scatter plots with correlation statistics for continuous variables.

5. Statistical Testing on Violin/Box Plots

Statistical testing has been implemented in this enhanced feature, supporting both global (kruskal-wallis or anova) and pairwise (wilcoxon or t.test) comparisons across groups in the “Violin / Boxplot” tab. Users can directly draw robust statistical conclusions within the app.

6. Flexible Ordering on Proportion Plots

The “Proportion plot” tab now allows flexible ordering of the X-axis. Users can sort the groups/samples by cell proportions, making it easier to identify clusters that are dominated by cells from specific biological or experimental conditions.

7. Customized Color Scale on Bubble Plot/Heatmap

The final enhanced feature offers customizable color scale in the “Bubbleplot / Heatmap” tab, through max value clipping. Users can now drag teh slider to adjust the dynamic range of the visualization and highlight the relevant data patterns.

— THE END —
Thank you for reading this tutorial :)